#610 — July 31, 2025 |
|
Postgres Weekly |
![]() |
Making Postgres 42,000x Slower (Because I Am Unemployed) — We run a lot of items on getting the most out of Postgres and making things run as quickly and smoothly as possible, but could you learn something by going the other way in an attempt to make Postgres as slow as possible? A seemingly pointless exercise, but a good way to learn about what some of Postgres’s configuration options actually do. Jacob Jackson |
💡 If you'd prefer something more directly usable, Jacob has also written a quick(ish) introduction to tuning Postgres. |
![]() Speed Up Postgres Queries With pganalyze Query Advisor — See how our new Query Advisor feature detected plan issues and sped up a Postgres query by over 1000x. Watch the webinar on demand to learn how to spot inefficient queries, get rewrite recommendations, & track improvements. pganalyze sponsor |
IN BRIEF:
|
Sharding Postgres at Network Speed — The lead developer of the PgDog transaction pooler / Postgres sharding tool shares the technical details of PgDog’s new, high speed logical replication based re-sharding feature. Lev Kokotov |
📄 Why Postgres Needs Better Connection Security Defaults – Understanding the difference between 📄 How CERN Powers Ground-Breaking Physics with TimescaleDB – A case study, but with some interesting technical tidbits from the heart of Europe’s science industry. TigerData 📄 Migration and Upgrades: Achieving Near Zero-Downtime in Postgres Sebastian Insausti 📄 Postgres CDC to Iceberg: Lessons from Real-World Data Pipelines Yingjun Wu (RisingWave) |
|
RELEASES: |
pg_meminfo: Postgres Memory Usage Stats for Linux — A Linux-only extension to inspect detailed per-backend memory usage using simple SQL queries. It surfaces metrics like Shaun Thomas |
pgsql-http 1.7: An HTTP Client for Postgres — “Wouldn’t it be nice to be able to write a trigger that called a web service? Either to get back a result, or to poke that service into refreshing itself against the new state of the database?” Paul Ramsey |
|